Adhesives for joining metals together, techniques for securely bonding a metal and a synthetic resin and the like are needed not only in the manufacture of industrial parts for automobiles, consumer electrical products, industrial devices and so forth but also in a wide variety of industrial fields including building materials and heavy construction materials. Many adhesives have been developed to this end, including some truly outstanding adhesives.
However, more rational joining methods that do not involve the use of an adhesive have also been studied conventionally. An example of this is a method in which a high-strength engineering resin is integrated, without the use of an adhesive, with a light metal, such as magnesium, aluminum or an alloy of these, with stainless steel or with another such iron alloy. For instance, the inventors proposed a method in which a resin portion is formed by injecting a thermoplastic resin onto a metal part inserted beforehand in an metallic mold for injection molding and this molded article is joined with the metal part at the same time (hereinafter abbreviated as “injection joining”).
The inventors proposed a manufacturing technique in which a polybutylene terephthalate resin (hereinafter referred to as “PBT”) or a polyphenylene sulfide resin (hereinafter referred to as “PPS”) is joined by injection joining to an aluminum alloy (see, for example, WO 2003-064150: Patent Document 1). The inventors also proposed a joining technique in which somewhat large holes are formed in an anodized film of an aluminum material and a synthetic resin penetrates into the holes and bonding is made (see, for example, WO 2004/055248: Patent Document 2).
The principle behind the injection joining proposed in Patent Document 1 is that an aluminum alloy is immersed in a weak aqueous solution of a water-soluble amine compound and the aluminum alloy is finely etched by the weak basicity of the aqueous solution. With this immersion, the adsorption of the amine compound molecules to the aluminum alloy surface occurs at the same time. The aluminum alloy thus treated is inserted into an metallic mold for injection molding and a molten thermoplastic resin is injected at high pressure.
At this point, the thermoplastic resin and the amine compound molecules adsorbed to the aluminum alloy surface come together, which generate heat or bring about a polymer breaking reaction. In parallel with this chemical reaction, the molten resin touches the aluminum alloy held at the low temperature of the mold to be quenched, thus crystallizes and solidifies (crystallization reaction). The chemical reaction and the crystallization reaction are considered to be in a competitive reaction relation for some reason and in this case the chemical reaction proceeds, while the crystallization reaction is suppressed. As a result, the viscosity of the resin does not rise sharply even though the resin is cooled by the aluminum alloy and the resin can work its way into the ultrafine recesses on the aluminum alloy surface.
Consequently, the aluminum alloy and the thermoplastic resin are securely joined, without the resin separating from the aluminum alloy surface. Specifically, when there is a chemical reaction that suppresses a crystallization reaction, strong injection joining is possible. Actually, it has been confirmed that PBT or PPS, which are capable of undergoing a chemical reaction with amine compounds, can be joined by injection joining with this aluminum alloy. The inventors called the mechanism of this injection joining the NMT (short for Nano Molding Technology) theory (hypothesis).
Also, although not the NMT theory, there is a known technique with which chemical etching is performed beforehand, then a metal part is inserted into the mold of an injection molding machine and injection molding is performed using a thermoplastic resin material (see, for example, Japanese Patent Application Laid-Open No. 2001-225352: Patent Document 3). This technique is somewhat unsophisticated as a joining method and its joining strength does not rise to the level of that produced with the NMT theory, but while the NMT theory was aimed only at aluminum alloys, the advocates of the NMT theory (the inventors of the present invention) also believed that a new joining technique should be developed for the injection joining of metals other than aluminum alloys.
As a result of development conducted for this purpose, the inventors arrived at a novel technique, called the “new NMT” theory. This holds that there are conditions under which injection joining is possible without the chemical adsorption of an amine compound to the surface of a metal part, that is, without the assistance of any special chemical reaction or any particular exothermic reaction. This new NMT theory is discussed below and it has been proven with many kinds of metal alloys.
At least the following conditions are required with an injection joining theory based on the new NMT theory. The first is that a hard, highly crystalline resin be used, that is, that PPS, PBT or an aromatic polyamide be used. Furthermore, these need be a composition that is further improved for the purposes of injection joining. Another condition is that the surface layer of the metal part inserted into the mold be tough and hard and have a specific surface configuration.
For instance, when a shaped magnesium alloy is used, corrosion resistance is low with a plain magnesium alloy covered with a natural oxidation layer. In this case, a surface covered with a hard, highly crystalline ceramic can be obtained by subjecting this alloy to a chemical conversion treatment to change the surface layer into a metal oxide, a metal carbide or a metal phosphide. The above-mentioned conditions can be met with a magnesium alloy part having such a ceramic surface layer and having a texturing of the micron order.
The following applies theoretically if a case is considered in which a shaped magnesium alloy that has undergone such a surface treatment is inserted into a metallic mold for injection molding. Since the mold and the inserted shaped magnesium alloy are held at a temperature that is lower by approximately 150° C. than the melting point of the resin, the injected resin is quenched as soon as it enters the flow path inside the mold and it is very likely that it will be under its melting point at the moment when it approaches the magnesium metal part.
When any kind of crystalline resin is rapidly cooled from a molten state to below its melting point, it does not crystallize right away and there is a period, albeit brief, in which it remains in a molten state under its melting point or, in other words, in a supercooled state. If the diameter of the recesses in the shaped magnesium alloy is relatively large (about 1 to 10 μm) and the depth of these recesses is up to about half of the diameter, then with some resin compositions, the resin will be able to work its way into the recesses within the limited time in which microcrystals are produced after supercooling.
Also, even if the number density of polymer microcrystal groups produced is still low, the resin will be able to work its way in if the recesses are large. This is because the size of the microcrystals having the form, at which some kind of ordered state occurred in the molecular chains from molecular chains that were moving irregularly, is considered to be a size of a few nanometers to 10 nm, estimating from a molecular model.
Consequently, while it is not really the case that microcrystals can easily penetrate into ultrafine recesses with a diameter of 10 nm, it is possible that some of the resin flow might poke into the recesses if the irregular surface had a period of a few dozen nanometers. Still, since countless microcrystals are generated at the same time, there is a sharp rise in the viscosity of the resin flow at the distal ends of the injected resin or where it touches the metal surface of the mold.
The upshot of this is that, when using a resin whose crystallization rate during quenching has been slowed with a special compound, if the period is 1 to 10 μm and the depth is about 0.5 to 5 μm (or half the period), the molten resin will be able to penetrate down to the bottom of those recesses and, if there should be fine texturing with a period of about 10 to 100 nm on the inner wall surfaces of these recesses, then some of the resin flow might poke into the recesses of the gaps between this fine texturing. It just so happens that an ultrafine textured surface with a period of 10 to 50 nm has been observed by electron microscope observation of the surface of a magnesium alloy that has undergone chemical conversion treatment and it has been confirmed that there is the above-mentioned fine surface structure.
In addition to magnesium alloys, when injection joining is performed on other metal parts having a surface with a similar configuration, the resin flow is able to penetrate to the bottom of large recesses on the micron order (that is, peaks and valleys with a period of 1 to 10 μm, with the height difference between these peaks and valleys being about half the period) and, if the resin flow is caught by the harder, fine peaks and valleys among these large recesses, then it can be surmised that when the resin is crystallized and solidified in these large recesses, it will probably be quite difficult for it to be pulled out.
Alloy parts of copper, titanium or steel were actually produced by performing etching and chemical treatment in an attempt to achieve such a configuration and, when a modified PPS resin was used in injection joining, a considerably strong joint was obtained as a result. The surface of the shaped alloy was subjected to oxidation or chemical conversion treatment to produce ceramic fine-crystal groups such as metal oxide or an amorphous layer, which served as hard and tough spikes. That is, the fine texturing serves as spikes in the large recesses and the resin that has solidified in the large recesses will not come loose even when a strong peeling force is exerted on the resin, so that as a result there is a strong joint between the shaped alloy and the resin.
The modified PPS resin, etc., mentioned above will now be discussed. In injection molding, the resin composition is quenched by injection from its molten state to a temperature below its melting point. If a resin composition is obtained that has the property whereby the crystallization rate during quenching is slowed, then the lengthened time before crystallization allows the resin to penetrate into the tiny recesses on the metal alloy part that is inserted into the mold, such as those discussed above, and this gives rise to a stronger joint. This is an important condition for a resin composition that is suited to injection joining.
Based on this line of thinking, the inventors discovered that if a shaped magnesium alloy or other metal alloy is chemically etched as discussed above and the surface layer is made to be a hard ceramic by a surface treatment such as chemical conversion treatment, then a hard crystalline resin with a special composition can be joined by injection joining to this and good joinability can be obtained (see WO 2008/069252: Patent Document 4, WO 2008/047811: Patent Document 5, WO 2008/078714: Patent Document 6, WO 2008/081933: Patent Document 7 or PCT/JP 2008/062945: Patent Document 8). Each of these patent documents corresponds to a type of metal but what they all have in common is the above-mentioned new NMT theory. In other words, it can be seen that the technology discussed in these patent documents shares a common theory that does not depend on the type of metal.
Substantially final conditions for the new NMT theory will be discussed now. First, regarding the metal alloy, it is a basic requirement that a chemical treatment be performed that suits the type of metal alloy, resulting in a surface that matches the following conditions (1) to (3):
(1) the surface has micron-order roughness, which is a textured face in which the period of texturing is 1 to 10 μm and the height difference is about half that period;
(2) the surface of the walls inside the recesses has fine texturing with a period of 10 to 500 nm and most preferably 50 to 80 nm; and
(3) the surface is covered with a thin layer of a hard ceramic phase and, more specifically, is covered with a thin layer of a metal oxide or metal phosphide that is environmentally stable.
The concept is simple: if we assume that a liquid resin composition penetrates into the recesses on the surface of this metal alloy and is solidified and hardened after penetration, the metal alloy base material and the solidified resin component will be joined extremely tightly.
The injection joining of a thermoplastic resin will now be explained based on this new NMT theory as follows. When a hard, highly crystalline thermoplastic resin composition whose crystallization and solidification rate during quenching has been slowed is injected, the resin composition injected into the metallic mold for injection molding remains in a supercooled liquid state for a while after being cooled to a temperature below its melting point. Consequently, if the metal alloy is inserted into the metallic mold for injection molding beforehand, the resin can easily work its way into the recesses mentioned in the condition (1).
The resin can also penetrate into the recesses of the fine texturing in the condition (2) to a certain extent, although not completely. After this, crystallization proceeds faster and solidification is reached, then the resin that has penetrated into and solidified in the recesses is caught by the fine texturing in the condition (2), while this fine texturing is extremely hard because of the condition (3), so the resin securely fastened as if it were spiked and cannot come loose from the recesses. This is the idea behind injection joining using a thermoplastic resin (Patent Documents 4 to 8).
The joining method can also be changed along the same concept as the new NMT theory. That is, a resin composition is produced beforehand by a method such as injection joining using as the raw material a resin composition whose main component is a hard, highly crystalline resin. Meanwhile, a metal alloy piece that satisfies the above conditions (1) to (3) is produced and this is heated with a hot plate or the like. The above-mentioned resin composition is pressed against this heated metal alloy piece. If the temperature of the metal alloy piece is higher than the melting point of the resin composition, the resin composition will melt at the contact surface.
If this product is allowed to stand and the temperature at the contact face between the metal and resin drops below the resin melting point over a period of a few seconds to 10 seconds or more, part of the molten resin will penetrate into the recesses on the metal surface, after which it will crystallize and solidify. If this method is used, there is no need for the crystallization and solidification rate during quenching to be slowed with a special compound or the like, so the conditions required by the resin composition are less severe. Unlike with injection joining, the pressure during penetration is extremely low, at only about 1 atmosphere, even if the environment during melting is returned to normal pressure from a vacuum. Thus, while it is impossible to raise the joining strength to the highest level, such a joining strength can be obtained that is usable from a practical standpoint. This is a molded article pressure welding method involving the use of a highly crystalline thermoplastic resin (see PCT/JP 2008/066009: Patent Document 9).
Also, if the joining mechanism of the new NMT theory is correct, then we can anticipate that bonding with a one-component thermosetting adhesive will yield extremely strong joining. Specifically, this is because it is surmised that the liquid resin approaches the metal alloy that has undergone surface treatment according to the new NMT theory and penetrates into the micron-ordered recesses, the resin penetrates to a certain extent into the gaps between the fine texturing on the surface of the inner walls of these recesses and the resin is then solidified and hardened, so that a spiking effect prevents the solidified resin from coming loose from the recesses and a strong joint is obtained. However, how far the resin can penetrate into the gaps of the fine texturing is determined by the viscosity of the liquid resin in that environment (pressure and temperature).
In this sense, the principle is that powerful adhesive strength is obtained by bonding with a one-component thermosetting adhesive, while the liquid viscosity of the resin in its unsolidified state is an important point. The inventors performed surface treatment of metal alloy pieces according to the new NMT theory, bonded two of these metal alloy pieces together using a one-component epoxy adhesive for general use and confirmed that powerful adhesion of 40 to 70 MPa was obtained in tests for shear breaking strength and tensile breaking strength.
Some slight modifications were made after adhesive coating. Specifically, the coated material was put in a desiccator, placed under a vacuum and then returned to normal pressure, thus the procedure was repeated. The pressure difference was only about 1 atmosphere but the liquid adhesive seems to readily penetrate into the recesses on the metal surface. After this, the coated metal alloy pieces were fixed together with clips or the like, heated and the adhesive was cured, which gave a securely bonded metal alloy pieces not seen in the past. The inventors have named this the NAT (short for Nano Adhesion Technology) theory, to make it clear that it is distinct from technology that utilizes injection molding.
The reason a one-component adhesive is preferable with the NAT theory is that such an adhesive will not gel in coating and subsequent pre-curing operations and the molecules that make up the resin component have a small molecular diameter, so this adhesive can penetrate to a certain extent into the gaps between the fine texturing in the condition (2). Even with a two-part thermosetting adhesive, joint strength will increase if a metal alloy is used that has undergone surface treatment according to the NAT theory but this will not usually rise to the level of a dramatic increase in adhesion strength. This is because nearly all two-part adhesives begin to gel the instant the curing agent component is added to and mixed with the main liquid and, if gelling occurs, there will be less penetration of the resin component into the gaps of the fine texturing in the condition (2).
In other words, when a two-part adhesive is used, adhesion strength often varies with the length of time that has passed after the curing agent was mixed in, which is undesirable in that it can result in inferior stability or inferior reproducibility. However, it should be understood that even an epoxy resin adhesive in which the curing agent is an acid anhydride, which is generally considered to be a two-part adhesive, can be used favorably if it takes a long time for gelling to start and if the gelling temperature is high. Adhesives such as this can be considered to be the same as a one-component adhesive.
The same can be said about a phenol resin-based adhesive or an unsaturated polyester resin-based adhesive. Specifically, there are commercially available phenol resin-based adhesives but most of them contain a solvent and are thus not solvent-free as are most epoxy adhesives. However, if the coating is allowed to stand for a while after application, so that the solvent volatilizes and the adhesive is cured, and if this product is then put under reduced pressure and returned to normal pressure at a medium temperature of 50 to 70° C., the phenol resin left behind after solvent volatilization will also melt and change into a viscous liquid with a viscosity of about 10 Pa seconds, so any air on the metal textured surface can be removed.
Although unsaturated polyester adhesives are not available on the market, there are many kinds of commercially available unsaturated polyester components that are used to produce glass fiber reinforced plastics (hereinafter referred to as GFRP) and organic peroxides that are admixed to these for curing under heating are also on the market. If the two are mixed in a suitable recipe, gelling and solidification will proceed as the temperature rises, without the gelling occurring right away, so such a blend can substantially be used as a one-component thermosetting adhesive.
The metal alloys discussed above are not the only adherends that can be bonded to a metal alloy. When a phenol resin-based adhesive is used as the adhesive, then friction materials and grinding materials comprising a matrix of phenol resin will also be bonded readily and when an epoxy adhesive is used for the adhesive, a carbon fiber reinforced plastic (hereinafter referred to as CFRP) comprising a matrix of epoxy resin will also be bonded readily.
Also, when an unsaturated polyester resin-based adhesive is used for the adhesive, a GFRP comprising a matrix of an unsaturated polyester resin will also be bonded readily. With all of these, a metal alloy piece that has been coated with an adhesive is brought into contact with a prepreg and fixed and these are heated and cured in this situation to solidify both the adhesive and the prepreg, allowing a composite to be obtained in which a metal alloy and a fiber reinforced plastic (hereinafter referred to as FRP) are securely bonded and integrated (see PCT/JP 2008/054539: Patent Document 10, PCT/JP 2008/057309: Patent Document 11, PCT/JP 2008/056820: Patent Document 12, PCT/JP 2008/057131: Patent Document 13, PCT/JP 2008/057922: Patent Document 14 or PCT/JP 2008/059783: Patent Document 15).
Similar technology related to injection joining that involves the use of a metal alloy and a thermoplastic resin was reported in the past (Patent Document 3). It will be stated here that the technology discussed in Patent Document 3 is not injection joining technology but rather technology that utilizes the relation between the linear expansion of a metal and the molding shrinkage of a resin.
As shown in Patent Document 3, if injection molding of a thermoplastic resin is performed around a peripheral component through which a metal rod-shaped substance has been passed, when the molded article is taken out of the mold and allowed to be cooled, the metal rod is fastened by the resin molded part. The reason for this is that the linear expansion of a metal such as an aluminum alloy, a magnesium alloy or a copper alloy is at most 1.7 to 2.5×105° C.−1 and, even if the metal is taken out of the mold and cooled to room temperature, the shrinkage is only 0.2 to 0.3% (linear expansion×about 100° C.)
However, the molding shrinkage of the resin is about 1% with PPS and 0.5% for glass fiber-containing PPS and, even for a resin in which the amount of filler has been increased, after injection molding the resin component will always shrink more than the metal portion. Therefore, if a shaped article in which there is a metal part in the middle and this part extends through the resin component is produced by injection molding with an insert, an integrated product can be manufactured with which the metal part is not likely to come loose because of the fastening effect of the molding shrinkage of the resin.
This method for manufacturing a fastening type of integrated article of metal and resin is known conventionally and the handle of a kerosene heater is an example of a similar molded article. A thick wire made of iron and having a diameter of about 2 mm is inserted into a metallic mold for injection molding and a heat-resistant resin or the like is injected. The wire has been given jagged marks (knurling) so that the resin will stay in place. Patent Document 3 is characterized by the fact that the method for texturing is changed from a physical method to a chemical method and is therefore neater, with the texturing being somewhat finer and that the gripping effect is enhanced by the generous use of a resin that is both hard and crystalline. Actually, it is stated in Patent Document 3 that gas leakage occurring along the metal rod is greatly suppressed by the disclosed technology but no mention is made of joining strength.
Meanwhile, with the inventions by the present inventors disclosed in Patent Document 1 and Patent Documents 4 to 8, there is no need at all for a resin fastening effect. A powerful force is necessary to break a molded article in which two flat pieces have been joined together. Another major feature of the technique for increasing joining strength according to the present invention is the use of a high-hardness crystalline resin composition that crystallizes and solidifies over a longer period of time during quenching.
To maintain a metal and a thermoplastic resin in a joined state stably over an extended period, it is actually necessary for the linear expansion coefficient of the two to be close to each other. The linear expansion coefficient of a thermoplastic resin composition can be lowered considerably by adding a large quantity of glass fiber, carbon fiber or another such reinforcing fiber, that is, filler.
Examples of using common steel materials are given in the above-mentioned Patent Documents 8 and 15. The inventors have subjected cold rolled steel, hot rolled steel and the like, which are the most common structural steel materials, to surface treatment so as to be suited to injection joining and adhesive joining or, in other words, suited to the new NMT theory and the NAT theory. This was all well in itself but some of the building material industry did not exhibit the expected interest in these inventions and the inventors were advised to develop something else in the first place for outdoor housing materials applications involving the envisioned integration of a GFRP material and steel.
Zinc-plated steel and aluminum-plated steel sheets are used in building material applications where the material needs to be maintenance free for 10 to 20 years and precoated steel sheets (also called colored steel sheets), in which one, two or in some cases three coats of baked-on finish are added to these steel sheets and so forth, have come to be used in large quantities because of their corrosion resistance and suitability to after-working. Also, because aluminum-plated steel sheets have excellent corrosion resistance at high temperatures, they are used not only in the above applications but also in heater covers, high-temperature gas exhaust pipes and the like. Because of this, the inventors conducted research and development into zinc-plated steel and aluminum-plated steel sheets in addition to ordinary steel materials.
At this time, the inventors conducted research and development to find out if the joining strength between an aluminum-plated steel sheet and an upper coating layer could be further improved by applying the NAT theory, if high adhesive strength could be obtained that was unattainable in the past with epoxy adhesives, phenol resin adhesives and unsaturated polyester resin adhesives that were the original object of the NAT theory, if injection joining would be possible with PBT and PPS resins according to the new NMT theory and so forth.
The present invention was the result of research and development that was important in terms of proving the above-mentioned NMT theory, new NMT theory and NAT theory, but the conclusion came as a surprise. Specifically, while the object of the present invention was completely attained, the surface configuration of the treated aluminum-plated steel sheet that achieved this object was different from the conditions required by the NMT theory, new NMT theory and NAT theory. This can be understood from FIGS. 13 and 14, which show the results of electron microscope observation. This hard-to-define surface configuration is the second one for the inventors, after an α-β titanium alloy.
However, although unlike the product by surface treatment according to the NMT theory and the new NMT theory of an extension-use aluminum alloy for which a simple photographic image is obtained, the large, white three-dimensional objects have a complex shape having an undercut structure. With the small texturing on the plain, if an amine compound has been adsorbed, the injected resin would probably penetrate into and securely solidify in the gaps formed by the recesses and the bases of the convex components, just as predicted by the NMT theory. Consequently, although different from the surface configuration defined in the NMT theory, new NMT theory and NAT theory, the result was configuration with something similar could be expected to be obtained.
Thus, what is currently wanted is related to a practical technique for obtaining a product in which a steel sheet and a resin are tightly integrated by injection joining of a thermoplastic resin composition to an aluminum-plated steel sheet material, by pressure welding of a molded thermoplastic resin composition to an aluminum-plated steel sheet material or by adhesively bonding either two aluminum-plated steel sheet materials together or an aluminum-plated steel sheet material and another adherend, as well as to further improvement of the corrosion resistance of an aluminum-plated steel sheet by coating an aluminum-plated steel sheet and obtaining a tightly bonded coating film.
If a composite could be manufactured in which an aluminum-plated steel sheet and a molded resin were tightly integrated, it could be used advantageously in building material-related parts that are mainly intended for use outdoors. Also, if a corrosion resistant coating layer could be strongly bonded to an aluminum-plated steel sheet, then not only would this help increase the corrosion resistance of the aluminum-plated steel sheet but it would also serve to improve the performance of a precoated steel sheet in which the aluminum-plated steel sheet was used.